Deconstructing ECRI's Defense of its Recession Call
Foreword from dshort: On Friday of last week, Lakshman Achuthan, the Co-founder of ECRI, returned to the media circuit defending ECRI's recession call, which I highlighted in my latest ECRI weekly update. Over the weekend Georg Vrba, whose indicator analysis has been featured at Advisor Perspectives, sent me an email with some pointed criticism of Achuthan comments on CNBC. He has given me permission to share with the Advisor Perspectives audience.
There were two incorrect statements that Lakshman Achuthan made at his TV appearance last Friday on CNBC. First he claimed that "GDP growth year-over-year fell down to 1.5% by Q2 2011." This statement is wrong. Check the GDP numbers and you will find that the GDP growth year over year fell to 1.63% by Q2 2011.
Second, and more worrisome is this Achuthan claim:
|The year-on-year growth of the Coincident Index, now at a 21-month low, has not had a decline like that in the past 50 years without a recession following in short order.|
This statement is also incorrect. The YoY growth of the ECRI Coincident Index has fallen from a high 3.70% to 1.94% over the last 66 weeks, which one can calculate from the Excel spreadsheet shared with the general public on the ECRI website. This is a decline of only 1.76%.
A far steeper fall was recorded from January 1995 to January 1996. The YoY growth of the ECRI Coincident Index fell then from a high of 5.23% to a low of 2.06% over 52 weeks. This is a decline of 3.17% to essentially the same 2.0% that the index has recently declined to, but in a much shorter time. No recession followed then, unless you want to apply this signal to the recession which followed in 2001 -- five years later.
Here is a chart, similar to the chart that Achuthan presented, highlighting the details.
This is a typical example of incorrect reporting to make the point the "expert" favors. See my recent Advisor Perspectives article, which shows similar false statements from David Rosenberg. It is time that the pundits take responsibility for what they are saying. I it certainly noteworthy that other indicators, like the Conference Board's LEI, the SuperIndex, etc. do not signal a recession now.
Here is another chart, the 6-mo annualized growth rate of the ECRI Coincident Index, which shows that the economy is actually improving. Obviously Achuthan did not show us this chart.
Why is everybody so hung-up about the WLIg when a much better recession predictor is the six-month annualized growth of ECRI US coincident Index. This indicator, when plotted in real-time (i.e. when the index is reported by ECRI instead of when it is listed) gets a good score of 0.362 from my evaluation system. You can see that it forecasted every recession and was never late in doing so. The only fault of this indicator is that after the recessions ended it still indicated recession when there was none, but in forecasting it gets an A+. So why is it ignored? Perhaps in our fast-moving world we prefer an indicator every week over one that only comes out once a month.
As anybody can see, it does not forecast recession now. I would rely on this indicator more than on Achathan's forecast.
Note: For a further discussion of the superiority of a six-month annualized growth indicator versus YoY, see the 1999 article published by Anirvan Banerji, the Chief Research Officer at ECRI: The three Ps: simple tools for monitoring economic cycles - pronounced, pervasive and persistent economic indicators.